# Hallucination Detection on WebQSP (Operational Metrics)

Source: [Hallucination Detection on WebQSP (Operational Metrics)](<https://www.sota2.com/research/sota/hallucination-detection-on-webqsp-operational-metrics>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-02T07:57:21.096Z

Evaluation of model performance and operational costs (LLM calls, total resources) for hallucination detection using the WebQSP dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Hallucination Detection |
| Dataset (including any stated split) | WebQSP |
| Metric | LLM Calls |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [LLM Calls](<https://www.sota2.com/research/sota/hallucination-detection-on-webqsp-operational-metrics.md?metric=LLM%20Calls>): Lower is better; unit: SCALAR.

- [Detector Size (M)](<https://www.sota2.com/research/sota/hallucination-detection-on-webqsp-operational-metrics.md?metric=Detector%20Size%20(M)>): Higher is better; unit: SIZE.

- [Total Resources (M/B)](<https://www.sota2.com/research/sota/hallucination-detection-on-webqsp-operational-metrics.md?metric=Total%20Resources%20(M%2FB)>): Lower is better; unit: SIZE.

- [Factor vs Ours](<https://www.sota2.com/research/sota/hallucination-detection-on-webqsp-operational-metrics.md?metric=Factor%20vs%20Ours>): Higher is better; unit: SCALAR.

## Recorded results

| Method | LLM Calls | Result date | Report paper |
| --- | --- | --- | --- |
| KG-Guard(GraphTransformer) | 0 | Not recorded | 6a1e63b3dbd7150506f85292 |
| KG-Guard(GAT) | 0 | Not recorded | 6a1e63b3dbd7150506f85292 |
| Llama 4 Scout | 1 | Not recorded | 6a1e63b3dbd7150506f85292 |
| GPT 5 Mini | 1 | Not recorded | 6a1e63b3dbd7150506f85292 |
| SelfCheck(N=5) | 5 | Not recorded | 6a1e63b3dbd7150506f85292 |
| SelfCheck(N=10) | 10 | Not recorded | 6a1e63b3dbd7150506f85292 |

### KG-Guard(GraphTransformer)

LLM Calls: 0. Result ID: 6a1e8cd7fbc373592c187c2b.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Encoder":"all-roberta-large-v1"}. This record does not assert independent reproduction.

### KG-Guard(GAT)

LLM Calls: 0. Result ID: 6a1e8cd7fbc373592c187c2c.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: {"Encoder":"all-roberta-large-v1"}. This record does not assert independent reproduction.

### Llama 4 Scout

LLM Calls: 1. Result ID: 6a1e8cd7fbc373592c187c2d.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT 5 Mini

LLM Calls: 1. Result ID: 6a1e8cd7fbc373592c187c30.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695f39f053fdd5c361743ecb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SelfCheck(N=5)

LLM Calls: 5. Result ID: 6a1e8cd7fbc373592c187c2f.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d943.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SelfCheck(N=10)

LLM Calls: 10. Result ID: 6a1e8cd7fbc373592c187c2e.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b3dbd7150506f85292.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d71e209131cea4d943.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
